The importance of determining the limit of detection of non-invasive prenatal testing methods.

OBJECTIVE Several non-invasive prenatal testing (NIPT) methods, which analyze circulating fetal cell-free DNA (cfDNA) in maternal plasma, suggest a fetal fraction (FF) ≥ 4% for a reportable result, with the assumption that fetal aneuploidies may not be detectable at lower FF. This study determined the actual limit of detection (LOD) of a massively parallel sequencing-based NIPT method and evaluated its performance in testing samples with low FF. METHOD An experimental model, involving the creation of artificial plasma mixtures with a final aneuploid FF ranging from 1% to 4%, simulated samples at different proportions of fetal cfDNA. We then analyzed 7103 blood samples, from pregnant women undergoing NIPT, to assess the impact of low FF on the performance of cfDNA testing. RESULTS Detection of common aneuploidies in samples with an FF as low as 2% is well within the ability of this technology. Of 105 pregnancies confirmed chromosomally abnormal, 25 (23.8%) involving a 2% <  FF  <  4% were consistently detected. These high-risk pregnancies would have not been identified using the suggested 4% FF cut-off. CONCLUSION This study underscores the importance of determining the actual LOD for each specific NIPT methodology. It may reduce the incidence of test cancelations and shorten the time required for the diagnosis of aneuploidy. © 2016 John Wiley & Sons, Ltd.